Use (char *) for fsname and subtype #20
Conversation
Attempt to fix #19
| {"offset=%zu", offsetof(sqfs_opts, offset), 0}, | ||
| {"timeout=%u", offsetof(sqfs_opts, idle_timeout_secs), 0}, | ||
| {"fsname=%s", (char *)"squashfuse", 0}, | ||
| {"subtype=%s", (char *)"squashfuse", 0}, | ||
| FUSE_OPT_END |
There was a problem hiding this comment.
According to the documentation of fuse_opt and the signature of fuse_opt_parse, the "squashfuse" strings should probably be a part of the struct passed to fuse_opt_parse.
There was a problem hiding this comment.
Not sure how to do this, would you mind sending a pull request?
There was a problem hiding this comment.
I can try, but I need some context. I found probonopd/static-tools#31, which suggests that fsname and fstype were added so that mount would print:
some.AppImage on /tmp... type fuse.squashfuse ...
instead of:
some.AppImage on /tmp/... type fuse.some.AppImage ...
Is that right? Currently, mount prints the latter.
There was a problem hiding this comment.
Yes, we want fuse.squashfuse.
There was a problem hiding this comment.
So this means that at least on Linux we shouldn't set fsname, because it will also change the first string (some.AppImage) to squashfuse, giving:
squashfuse on /tmp/... type fuse.squashfuse ...
But I'm not sure about FreeBSD, because it seems that it doesn't have the subtype option (see AppImage/AppImageKit#1139 and https://github.com/libfuse/libfuse/blob/a466241b45d1dd0bf685513bdeefd6448b63beb6/lib/helper.c#L195).
There was a problem hiding this comment.
On FreeBSD, fuse is ok (it is the same for other fuse filesystems).
There was a problem hiding this comment.
Or has this been fixed in freebsd/freebsd-src@128a1db806d?
There was a problem hiding this comment.
Looks like setting only subtype=squashfuse implicitly sets fsname; mount gives:
squashfuse on /tmp/... type fuse.squashfuse ...
So if we want to keep the AppImage name as the filesystem source (i.e. some.AppImage /tmp/... type fuse.squashfuse), we should set it explicitly via fsname.
Attempt to fix #19